m3soulu / light-oauth2 Goto Github PK

Should we implement this and does it affect other tests?

Make a list of all error tags to be used in the data collection loop.

Should be possible by using regex or grep.

Some tasks are still not migrated to use get__name__(), success etc

Currently, locust output summary has lines like the follwing:

GET      /oauth2/user/5ab548a0                                    3      3      3      3      3      3      3      3      3      3      3      2
DELETE   /oauth2/user/5f8978aa                                    7      7      7      7      7      7      7      7      7      7      7      1
DELETE   /oauth2/user/6073d749                                    8      8      8      8      8      8      8      8      8      8      8      1
DELETE   /oauth2/user/6108d7d6                                    6      6      6      6      6      6      6      6      6      6      6      1
DELETE   /oauth2/user/687e3620                                    6      6      6      6      6      6      6      6      6      6      6      1
GET      /oauth2/user/687e3620                                    8      8      8      8      8      8      8      8      8      

because we are inserting parameters into the URL.

It is possible to use the name parameter in the request to modify what string is used to group requests.
https://docs.locust.io/en/stable/writing-a-locustfile.html#grouping-requests

Some tests do not behave as expected due to errors on the light-oauth2 side.

We should find all of them and document here with label light-oauth error.

Should we implement this, and does it affect other tests?

Should this be implemented and does it affect other tests?

For some reason, the mysql container does not automatically read the file put into it during deployment, even though it is supposed to according to documentation.

Restarting the containers also requires logging into the mysql container and manually piping the file into mysql process.

Should we implement this and does it affect other tests?

Do we want to redeploy the docker containers (or at least refresh the database) when executing different errors?

If so, we need to account for #1